Fractionated Doses of Gemtuzumab Ozogamicin (GO) Combined to Standard Chemotherapy (CT) Improve Event-Free and Overall Survival in Newly-Diagnosed De Novo AML Patients Aged 50–70 Years Old: A Prospective Randomized Phase 3 Trial From the Acute Leukemia French Association (ALFA)

GO is a potent antibody-directed chemotherapy against CD33 antigen. Two MRC and SWOG Phase 3 studies have compared standard CT alone or combined with one single GO infusion (at 3 and 6 mg/m², respectively) in younger adults with AML with contradictory results (Burnett, JCO 2011; Petersdof, Blood 2009). We have shown in relapsed AML Phase 2 studies that fractionated infusion of GO 3 mg/m² on day 1, 4 and 7 was effective and might be safely combined to standard 3+7 DNR/AraC induction (Taksin, Leukemia 2007; Farhat, AJH, accepted). Here, we report the results of the prospective open label randomized multicentric Phase 3 ALFA 0701 trial (ClinicalTrial.gov ID, NCT00927498) designed to evaluate the efficacy and safety of adding this fractionated GO schedule to standard front-line chemotherapy in older AML pts.

Eligible patients (pts) were adults aged 50–70 years old with previously untreated de novo AML. Pts were randomized to receive induction with DNR 60 mg/m²/d on day 1–3 and AraC 200 mg/m²/d CI on day 1–7, without (DA arm) or with GO at 3 mg/m²/d on day 1, 4 and 7(DAGO arm). Pts with persistent marrow blasts at day 15 received additional DNR 35 mg/m²/d on day 1–2 and AraC 1g/m²/12h on day 1–3. Pts achieving CR/CRp received two consolidation courses with DNR 60 mg/m²/d on day 1 and AraC 1 g/m²/12h on day 1–4, ± GO at 3 mg/m²/d on day 1 according to the randomization arm. The primary study objective was event-free survival (EFS). The study was designed to detect a 25% to 40% EFS gain at 3 years, (two-sided test, power 80%, type I error 5%). Secondary objectives were response rate, disease-free survival (DFS), overall survival (OS), and safety.

From March 2008 to November 2010, the required sample of 280 pts (median age, 62 years) was enrolled. Nine pts did not satisfy for inclusion criteria and were excluded from analysis. Cytogenetics was favorable (N=9), intermediate (N=177), adverse (N= 57), not done/failure (N=28). Overall, 52 pts had a favorable NPM1+ w/o FLT3-ITD genotype. The two treatment arms were well matched for all pre-treatment characteristics including age, sex, ECOG-PS, WBC, cytogenetics and molecular characteristics. CR+CRp was achieved in 220/271 pts (77%): 100/134 (75%) in the control DA arm versus 110/137 (80%) in the DAGO arm (P=0.31). There were 5/134 (4%) induction deaths in DA arm and 9/137 (6%) in DAGO arm (P=0.41). Primary resistant AML rate was 29/134 (22%) after DA versus 18/137 (13%) after DAGO (P=0.08). At 2 years, EFS was estimated at 15.6% in DA arm versus 41.4% in the DAGO arm (P=0.0018), while DFS was 18.1% in DA arm and 48.5% in the DAGO arm (P=0.0009). This significant benefit in EFS (primary objective) was observed in pts aged <65 years (P=0.035) as well as in older pts (P=0.019), and persisted after censoring the 39 pts who received allogeneic stem cell transplantation in first CR/CRp at transplant time (P=0.015). Subgroup analysis showed that EFS benefit persisted in pts with favorable/intermediate karyotype (P=0.0008) while not in those with adverse karyotype (P=0.25). Interestingly, EFS benefit was still observed when excluding favorable pts (favorable karyotype or genotype) from the comparison (P=0.0017). Finally, in the whole patient population, this gain in EFS translated into a longer OS (median, 25.4 versus 15.3 months in DAGO versus DA pts; P=0.037). Besides treatment arm, cytogenetics and favorable genotype were the only factors predictive of outcome. After adjustment on these factors, DAGO treatment remained significantly associated with longer EFS (P=0.009), DFS (P=0.003), but not OS (P=0.14). The rate of fatal adverse events at least possibly attributable to treatment was 9/134 (6.7%) in the DA and 12/137 (8.7%) in the DAGO arm (P=0.65). Prolonged grade ≥ 3 thrombocytopenia was observed in 19 DAGO pts, either after induction (N= 4) or first consolidation (N=15). Three liver VOD were observed in the DAGO arm (2 during induction, 1 during first consolidation), 2 being associated with a fatal issue. No difference was observed between both arms in the incidence of severe sepsis (DAGO 18.9%, DA 14%), as well as in the rate of intensive care unit admission during the course of therapy (DAGO 14.5%, DA 12.6%).

The addition of fractionated doses of GO (3 mg/m²/d on day 1, 4, and 7) to standard CT significantly improves EFS and to a less degree OS in AML pts aged 50–70 years old. The main toxicity observed with GO was prolonged thrombocytopenia in 19 patients and 3 episodes of VOD.

Castaigne:Pfizer/Wyeth: Consultancy, Honoraria, Membership on an entity’s Board of Directors or advisory committees. Off Label Use: Gemtuzumab Ozogamicin is available in Europe as a compassionate treatment for relapsed AML. In this study patients GO was used in front-line treatment.